Re: [platform-swt-dev] GTK2: warnings and criticals

[Seth Nickell <snickell@xxxxxxxxxxxx>]

Hi Boris,

Its not different, I just didn't find logProc. Thanks!

-Seth

On Fri, 2002-08-23 at 15:11, Boris Shingarov wrote:
> Seth,
> 
> I don't understand how this functionality is different from what we are 
> doing in Device.logProc() and friends?
> 
> 
> 
> 
> 
> Seth Nickell <snickell@xxxxxxxxxxxx>
> Sent by: platform-swt-dev-admin@xxxxxxxxxxx
> 08/23/2002 01:18 PM
> Please respond to platform-swt-dev
> 
>  
>         To:     platform-swt-dev@xxxxxxxxxxx
>         cc: 
>         Subject:        [platform-swt-dev] GTK2: warnings and criticals
> 
> 
> We're currently just letting GTK warnings and criticals go to the
> console. I've written a little code that intercepts warnings and
> criticals (we could intercept other messages too, but these are the most
> common ones triggered by misuse of GTK), and passes them back to a
> static function in OS.java which could then trigger appropriate
> exceptions (or just break in Java). This should make finding bugs in
> GTK/SWT code much easier since you can trivially track down the call
> that caused the warning.
> 
> I also have a code snippet along the same lines that causes warnings and
> criticals to break in a C debugger (both this and the Java stuff can be
> used in conjunction with one another). This allows you to attach gdb to
> a process being run which will hit a breakpoint whenever GTK throws a
> warning. Sometimes it is easier to figure out why a warning was being
> thrown from the C side rather than the Java side.
> 
> Code isn't polished or finished, but it helps me out a lot so I hope
> somebody else will find it useful too (and perhaps somebody will be
> interested in doing the work to integrate it with eclipse so we get java
> exceptions off the Gtk equivalent of exceptions).
> 
> -Seth
> 
> 
>